Session: 17-01-01: Research Posters

Paper Number: 150747

150747 - Metal Additive Manufacturing - Machine Learning Analysis and Inferences on the Process Parameters and Material Characteristic 

Selective Laser Melting (SLM) has emerged as a pivotal technique in additive manufacturing, particularly for the fabrication of metal components with complex geometries with high precision. This work not only explores the intricate relationship between SLM process parameters and the resulting material characteristics but also showcases the potential of machine learning (ML) techniques in optimizing the performance and reliability of manufactured parts.

The application of machine learning (ML) to Selective Laser Melting (SLM) in additive manufacturing demonstrates significant potential in optimizing process parameters and understanding material characteristics. Prior research has indicated that these parameters significantly affect the microstructure and mechanical properties of the final products. Thus, understanding their impacts and interactions is crucial for optimizing the SLM settings for various materials and applications. Critical process parameters, including laser power, scanning speed, and hatch spacing, were systematically designed using DOE (Design of experiment) to investigate their influence on printed samples' density, mechanical properties, and surface finish. Using experimental trials and advanced characterization techniques, such as scanning electron microscopy (SEM), micro-CT scan, and mechanical testing, we try to explain the physics of the material characteristics in the SLM process. SEM is utilized to examine the surface morphology and microstructure of specimens. This technique offers high-resolution imaging that can reveal detailed features such as the melt pool boundaries, the presence of porosities, and the texture of the printed layers. Micro-CT offers a non-destructive means to visualize and quantify the internal structure of SLM-fabricated components. This technique is particularly valuable for detecting and analyzing defects such as voids and inclusions within the bulk material, which are not visible on the surface.

In parallel, a machine learning model was developed and trained using experimental data to predict material characteristics based on SLM parameters. The ML model employed supervised learning algorithms to analyze experimental datasets and identify patterns correlating process parameters with material outcomes. This predictive capability enables more efficient process optimization by reducing the need for extensive trial-and-error experimentation.

Our findings highlight the pivotal role of process optimization in achieving desired material properties. This understanding is crucial in the field of additive manufacturing. Integrating machine learning into the SLM process provides a powerful tool for predicting and optimizing material characteristics. This can significantly enhance the parameter selection process in Selective Laser Melting (SLM) by providing optimized, data-driven solutions that improve the quality and efficiency of the manufacturing process. This approach accelerates the development of high-performance components and contributes to more sustainable and cost-effective manufacturing practices. Also, it provides valuable insights into optimizing SLM process parameters to tailor the microstructure and properties of metal components.

Presenting Author: Ram Mohan NC A&T State University

Presenting Author Biography: Dr. Ram Mohan is Professor of Nanoengineering at the Joint School of Nanoscience and Nanoengineering (JSNN), at North Carolina A&T State University, Greensboro, NC, USA. He also serves as an Adjunct Professor of Nanoscience at JSNN. Dr. Mohan is also an affiliated faculty with the computational science and engineering graduate program and serves as the co-lead of the computational modeling research cluster at North Carolina A & T State University.

Authors:

Nikhil Ingle NC A&T State University
Ram Mohan NC A&T State University